Sorry, using the iPhone app to make comments is not easy. Cluster failover is possible at any time and your report is welcome and I am sure it is a great fun project that brings you closer to 100% uptime if nothing else[;)] I was commenting more on pings starting and stopping mysteriously. Also I didn't think the OP was clustering his astaro.
I do believe that you are right and I"m certainally not trying to take over the thread lol. I just thought that it was ironic that here I am talking about that I've had no issues with mine as of yet and then boom what do you know. But my hat goes off to the peeps at Astaro because the failover is awesome. I even tested it while making a phone call through my asterisk box, there was about a 3 sec silence in the phone call and then it just picked right back up as if nothing happened.
I would love to hear more about possible ESXi misconfiguration. Any variables that relate I would like to stay on top of. I found the allow icmp checkbox and i can ping google again. I think in the first few minutes of setup ( when i pinged successfully before ) the rules might not be firmly in place.
I am very concerned with high availability at home as I run several web servers, public and private ( not to mention my whiny tenants that play WoW ) and will be using many of the services when I'm onsite with a client. Im leaning more towards running Astaro on its own box for now.
Feel free to PM me and I can help you with the HA setup. My current setup in ESXi is pretty vanilla. I told ESXi to just setup the machine as a Linux host for SUSE Enterprise 11 with 2GB of ram and 4 NICS, 1 for Internal, 1 External, 1 DMZ and 1 for HA. Other than that just a normal VM setup
High availability at home is pointless. Specially if you are already virtualizing. This beta has been fairly stable and not saying kernel panics are impossible but probably points to a misconfiguration in your esxi. I personally have never seen any of my systems just "laying down" on ANY version or beta I have ever used physical or virtual.
A daemon misbehaving.... sure. Linux kernel laying down, not likely without faulty hardware.
I have. I ahve one 8.3xx machine that's bombed inside of hyper-v..i then had a physical machine the same day bomb as well. Support and i are trying to figure out what is going on.
v9 in hyper-v has not connectivity at all for me right now.
Interesting.... that is why a large test base is always a good thing. I have been running mine on esxi 4.x since v8 first came out and install betas and usually leave my stable release running along side. So far v9 has been chugging along just fine even though I exclusively use vmxnet3 adapters for all my astaro builds.
yeah it's strange. i think they have a bug that's hard to find in hyper-v drivers...i've tried everything i can think of to get v9 beta to cooperate in hyper-v it just doesn't want to play nice(hyper-v and v9)
